Pith. sign in

REVIEW 2 major objections 5 minor 26 references

Accuracy analysis and optimization of scale-independent third-order WENO-Z scheme with critical-point accuracy preservation

T0 review · 2 major / 5 minor · reviewed 2026-08-04 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper claims that WENO3-ZES4, a scale-independent third-order WENO-Z variant, recovers full third-order accuracy at first-order critical points (CP1) wherever they fall inside the stencil, while using the exponent p=1 in its nonlinear w

desk verdict A solid incremental WENO3-Z variant whose central accuracy lemma holds up; the printed algebra needs fixing and the resolution claims carry a tuning caveat. read the letter →

arxiv 2509.08413 v1 pith:H6JWM3FZ submitted 2025-09-10 math.NA cs.NAphysics.comp-ph

Yunchuan Wu , Xiuzheng Cheng , Yi Duan , Linsen Zhang , Qin Li , Pan Yan , Mengyu Wang This is my paper · ORCID
classification math.NAcs.NAphysics.comp-ph MSC 65M0665M12
keywords WENO-Zschemecriticalpointaccuracysmoothnessindicatororderpreservationscaleindependencenonlinearweightsthird-orderresolutionoptimization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to remove a known weakness of the standard third-order WENO-Z shock-capturing scheme: when the solution has a first-order critical point (derivative zero, second and third derivatives nonzero), the scheme loses its third-order accuracy, and earlier fixes only work when the critical point lands on a grid node, are scale-dependent, and use large exponents that blur resolution. The core move is an accuracy-optimization lemma: if the two local smoothness indicators have identical leading error terms and differ only at the next order, the nonlinear weights gain an extra order of accuracy, enough to satisfy the optimal-order condition with exponent p=1. Using that lemma, the paper constructs new local and global smoothness indicators on a five-point stencil and fixes the free coefficients by balancing resolution against robustness, producing WENO3-ZES4. If right, the scheme holds third-order L-infinity convergence at a first-order critical point no matter where it sits in the interval, remains scale-independent, and resolves fine features that typical third-order schemes smear, such as the second density peak of the Shu-Osher problem on just 240 points.

What carries the argument

Lemma 3.1 (the accuracy-optimization lemma): for alpha_k = d_k(1 + C_alpha (tau/beta_k)^p), if the local smoothness indicators share identical coefficients on all error terms up to order Delta x^(n2-1) and differ only in the Delta x^n2 term, and C_alpha is common across stencils, then omega_k - d_k = O(Delta x^(p(m-n1)) times Delta x^(n2-n1)); the second factor is what permits p=1. The construction satisfying its hypotheses is the pair of local indicators beta_0*, beta_1* of Eqs. (17)-(18) on the five-point stencil {x_(j-2),...,x_(j+2)}, each combining a two-point first-derivative difference with a three-point second-derivative difference, with equal coefficients C_beta0 = C_beta1 = 2.0 so t

What would settle it

Run the paper's scalar advection test (initial condition Eq. (22), CFL 0.25, L-infinity error at t=2) with the moving first-order critical point forced to sit exactly at a half-node (lambda=1/2) on final refinement levels; the claim predicts a clean rate of 3.000. A stronger check: repeat with the critical point at other lambda values (e.g., 1/3, 0.75) and directly measure omega_k - d_k at the critical point; if the O(Delta x^2) scaling fails for some lambda, or if letting f''(x_c) approach zero drops the L-infinity rate below 3, the uniformly-at-third-order claim fails.

Watch

Extended reading notes

Core claim

The paper claims that the accuracy of WENO-type nonlinear weights can be upgraded without raising the exponent p. Lemma 3.1 states that when local smoothness indicators expand as beta_k = sum a_l' Delta x^l' + b_k Delta x^n2 + O(Delta x^(n2+1)) with common coefficients a_l' and differing b_k, and when the constants C_alpha in all non-normalized weights are equal, then omega_k - d_k = O(Delta x^(p(m-n1))) times O(Delta x^(n2-n1)); the second factor is a bonus order. Applied to third-order WENO-Z: with the five-point local indicators of Eqs. (17)-(18) set to equal C_beta = 2.0, beta_k* behave as (lambda^2 + C_beta) f''^2 Delta x^4 + b_k Delta x^5 + ..., so n1=4, n2=5 and the Delta x^4 coeffici

Load-bearing premise

The accuracy proof assumes the solution is smooth around the critical point and that both f''(x_c) and f'''(x_c) are nonzero, so the Taylor expansions of the two local indicators share the same Delta x^4 coefficient; if a discontinuity, a near-zero second derivative, or a critical point outside the five-point stencil enters, the p=1 accuracy upgrade is not established.

Editorial extensions

If this is right

  • Third-order WENO schemes can hold full accuracy at first-order critical points with exponent p=1, removing the resolution loss that large p caused in earlier fixes; the Shu-Osher second peak and valley are resolved with only 240 points.
  • Lemma 3.1 gives a transferable recipe: matching the lower-order error coefficients of local smoothness indicators buys n2-n1 extra orders of accuracy for the weights, relaxing how accurate the global indicator tau must be.
  • Scale-independence is kept: unlike WENO-NP3/-F3/-NN3/-PZ3, the new weights are invariant under rescaling of the variables, so the same computation gives the same answer at any scale.
  • Using a tau of only O(Delta x^5) at CP1 avoids the robustness problems that the higher-order indicators of WENO3-ZM and WENO3-ZES caused; the scheme completed every problem in its test battery, including the Mach-2000 jet and viscous double-cone flows.
  • The optimal-order condition is met for every lambda in (-1,1), so third-order L-infinity convergence holds whether the critical point is on a node, a half-node, or anywhere else in the interval.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Nothing in Lemma 3.1 is specific to third order; the same error-coefficient matching could lower the exponent for fifth-order WENO-Z at critical points, trading the usual p=2 for p=1 and sharper resolution, a direction the paper leaves untested.
  • Because the accuracy gain in Lemma 3.1 holds for any common value of C_beta, the third-order claim is insensitive to the Shu-Osher-calibrated parameters C_beta=2.0 and C_alpha=1.3; only resolution and robustness depend on them, so order tests away from the calibration point should still show the optimal rate.
  • The paper notes that b_k vanishes at isolated lambda values, which should produce local super-convergence; a systematic order-versus-lambda scan could locate the true worst case and possibly reveal simpler indicators satisfying Eq. (12).
  • For the Euler and Navier-Stokes tests the scalar proof carries over only through the local characteristic decomposition used in practice; the 'any position' guarantee for systems and for flows with discontinuities is empirical, not a proven extension.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 5 minor

Summary. The manuscript proposes a new third-order WENO-Z scheme, WENO3-ZES4, designed to retain third-order accuracy when a first-order critical point (CP1) occurs at an arbitrary position inside the stencil. The main theoretical contribution is Lemma 3.1, an 'accuracy-optimization lemma' stating that if the local smoothness indicators β_k share common lower-order Taylor coefficients and differ only at order n2, then normalized nonlinear weights with p=1 can gain an extra order of accuracy. The authors construct extended local indicators β_0^*, β_1^* (Eqs. (17)-(18)) and a global indicator τ_4 (Eq. (21)) so that at CP1 the required condition holds, and choose the free parameters C_β=2.0 and C_α=1.3 by a combination of Lemma 3.2 and numerical resolution tests. The paper reports a scalar advection convergence test showing third-order L∞ convergence, and a range of 1D and 2D Euler and Navier-Stokes validation cases arguing for good resolution and robustness.

Significance. If the central claim is correct, WENO3-ZES4 is a meaningful practical contribution: it is scale-independent, uses the favorable exponent p=1, and preserves third-order accuracy at CP1 for arbitrary critical-point position. The idea of upgrading the accuracy of nonlinear weights by exploiting a common leading-order part of β_k is interesting and goes beyond the standard WENO-Z analysis. The numerical validation is extensive and, where quantitative (Table 2), supports third-order convergence. The paper also ships explicit formulas and parameters, which is helpful for reproducibility. However, the proof machinery as printed is not fully trustworthy: the key expansion coefficient b_k in Eqs. (19)-(20) is incorrect, and the appendix proof of Lemma 3.1 contains garbled algebraic steps. These are localized and fixable, but they must be corrected before the central theoretical claim can be taken as established.

major comments (2)
  1. [§3.2.1, Eqs. (19)-(20)] The printed coefficient b_k = -4λ^3 + (2/3)λ - 2Cβλ ± 2Cβ in the paragraph after Eq. (20) does not follow from a direct Taylor expansion of Eqs. (17)-(18) about x_c = x_j + λΔx. Expanding β_0^* and β_1^* with f'=0, f''≠0, f'''≠0 gives β_k^* = (λ^2 + Cβ) f''^2 Δx^4 + b_k Δx^5 + O(Δx^6), with b_0 = (-λ^3 + (2/3 - 2Cβ)λ - 2Cβ) f''f''' and b_1 = (-λ^3 + (2/3 - 2Cβ)λ + 2Cβ) f''f'''. Thus b_1 - b_0 = 4Cβ f''f''', which is nonzero for all λ∈(-1,1) provided Cβ>0. So the qualitative conclusion of Lemma 3.1 still holds, but the printed coefficient is wrong and the factor f''f''' is missing. Please correct the expansion and state explicitly that b_1−b_0 ≠ 0 for all λ∈(-1,1).
  2. [Appendix, proof of Lemma 3.1] The proof contains an algebraic non-sequitur after multiplying numerator and denominator by (β_k/τ)^p. The numerator should be (β_k/τ)^p + C_{αk}, but the manuscript displays an expression with (τ/(τ β_k/β_k))^p, which is undefined and obscures the argument. The later application of a/(a+x) = 1 − x/a + O(x^2) is also not fully explicit: the 'small' quantity is x = −C_{αk} + Σ_l d_l C_{αl} + O(Δx^{n2−n1}), and the reader must verify that all O-terms are uniform. The proof is salvageable, and the lemma itself appears correct, but the printed derivation is not. Please rewrite the proof cleanly with all expansions explicitly justified.
minor comments (5)
  1. [§4.2.1, Table 2] The text says the CP1 initially at x=0 migrates to a half-node every four iterations, but at the final time t=2 it is again at a grid node. The second CP at x=-2+2x_c remains at a non-node at t=2, so the reported L∞ rates do test off-node behavior, but the paper should state this explicitly or report the L∞ error at an intermediate time when the x=0 CP is at a half-node.
  2. [§2.2, Lemmas 2.1 and 2.2] Lemmas 2.1 and 2.2 are imported from Ref. [13] without proof. Since Lemma 2.1(2) motivates stencil extension, please state explicitly that the proofs are given in [13], or include them for self-containedness.
  3. [§3.2.1, Eqs. (19)-(20)] The indexing b_1, b_2 in Eqs. (19)-(20) conflicts with the later b_k notation. Use b_0 and b_1 consistently for β_0^* and β_1^*.
  4. [Lemma 3.2] The notation δ_j^{(n)m} is introduced informally in §2.3 and used in Lemma 3.2. Please define it precisely, including whether m denotes order of accuracy and how the δ quantities relate to finite differences.
  5. [General] There are several typos, e.g., 'coherently free coefficient' in §2.1, 'gird' in Fig. 10 caption, and inconsistent use of Δx vs dx in Eqs. (19)-(20). A careful proofreading pass is recommended.

Circularity Check

2 steps flagged · score 4.0 of 10

Central third-order-at-CP1 derivation is self-contained, but the Shu–Osher resolution "validation" is fitted on the same test problem, and a self-cited uniqueness lemma is used to force the extended-stencil family.

  1. fitted input called prediction [Sec. 3.2.2 (parameter determination) and Sec. 4.2.2(1) Shu–Osher validation]
    "Specifically, we use the Shu–Osher problem, which is a canonical indicator for numerical resolution (as detailed in Section 4). Through comparative numerical studies, the optimal values of 𝐶𝛽 and 𝐶𝛼 are determined. ... For the Shu–Osher problem, typical schemes barely resolve the second peak, making its resolution a pivotal criterion for parameter determination. As a tradeoff between resolution and robustness, we use 𝐶𝛽 = 2.0 and 𝐶𝛼 = 1.3. Notably, WENO3 -ZES4 successfully resolves the secondary peak (marked by the dashed box), demonstrating its superior resolution."

    Cβ and Cα are free parameters of the scheme; they were selected by optimizing the density profile of the Shu–Osher problem on 240 grid points, with the second peak/valley as the criterion. The later "validating test" of the same problem on the same grid (Fig. 2, Sec. 4.2.2) reports the tuning objective, not an independent prediction. The resolution claim on Shu–Osher is therefore a fitted input renamed as a validation result.

  2. uniqueness imported from authors [Secs. 2.2–2.3, Lemma 2.1(2) and ensuing conclusion]
    "To construct an appropriate global IS, we proved the following lemma in [13]. Lemma 2.1: ... (2) With CP1 (𝑓′ = 0, 𝑓′′ and 𝑓′′′ ≠ 0) occurring at 𝑥𝑐 = 𝑥𝑗 + 𝜆 ∙ ∆𝑥 (−1 < 𝜆 < 1), there exist no non-trivial solutions 𝑎𝑖1,𝑖2 such that 𝜏(𝑓) expands to 𝑂(∆𝑥5) at the critical point. ... From Lemma 2.1(2), achieving optimal order recovery at critical points requires either extending the stencil to construct scale-independent schemes..."

    The impossibility/uniqueness result that rules out a three-point scale-independent τ of O(Δx^5) at CP1 is not proved in this paper; it is cited from Ref. [13] by the same group. It is then used to declare that stencil extension is required, steering the design to the authors' own extended-stencil family (WENO3-ZM, -ZES, -ZES2/3, -ZES4). The forced-choice premise thus rests on a self-citation that is not re-verified here.

full rationale

Most of the derivation chain is non-circular. Lemma 3.1 is stated and proved in the Appendix from explicit assumptions; the local indicators (17)–(18) and global indicator (21) are explicit formulas; the CP1 expansions (19)–(20) and the τ4 error are supplied; and the sufficient condition Eq. (7) is standard. Thus the central order claim—ω_k−d_k=O(Δx^2), giving third-order accuracy at a CP1 located anywhere in the five-point stencil with p=1—does not reduce to fitting or to the tuned constants: it only requires Cβ0=Cβ1 and a common Cα, not their numerical values. The two circular/self-citation elements are secondary. First, the resolution validation on the Shu–Osher problem is not independent: Cβ=2.0 and Cα=1.3 were selected by comparing density profiles on that very problem, with second-peak resolution as the criterion; Section 4.2.2 then reports success on the same problem as a validating test. Second, Lemma 2.1 (uniqueness/no non-trivial τ of O(Δx^5) on a three-point stencil) is imported from the authors' Ref. [13] and used to declare that stencil extension is required—a load-bearing self-citation for the scheme-family choice. These do not corrupt the accuracy proof, so the score is moderate rather than high. Separately, algebraic slips in the Appendix and in the printed b_k coefficients, and the fact that the L∞ table at t=2 samples CP1 at a grid node, are verifiability/coverage concerns rather than circularity.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The central order-preservation result is not heavily parameterized: any positive equal C_beta and any positive equal C_alpha should preserve the formal order. However, the resolution claims are parameterized by C_beta = 2.0 and C_alpha = 1.3, tuned on one benchmark. The paper also imports Lemmas 2.1 and 2.2 from the same group's earlier papers without proof. No new physical entities are introduced.

free parameters (2)
  • C_beta (C_beta0 = C_beta1) = 2.0
    Constant added to local smoothness indicators in Eqs. (17)-(18). Equality across the two stencils is required by Lemma 3.1, but the numerical value 2.0 is chosen by numerical trials on the Shu-Osher problem in Section 3.2.2.
  • C_alpha = 1.3
    Multiplier in nonlinear weights Eq. (5). Equality across the two weights is required by Lemma 3.1, but the magnitude 1.3 is tuned via the Shu-Osher problem in Section 3.2.2.
assumptions (4)
  • domain assumption CP1 is defined by f' = 0 with f'' and f''' nonzero, and Taylor expansions around x_c inside the stencil are valid.
    Used in Section 3.2.1 to derive the error orders of beta_0^*, beta_1^*, and tau4. If a shock or near-zero derivative sits in the stencil, these orders fail.
  • standard math The sufficient condition omega_k - d_k = O(Delta x^r) guarantees optimal order for WENO reconstruction.
    Stated in Section 2.2 via Eq. (7) and attributed to Refs. [3,4]; all later order conclusions use this condition without re-proving it.
  • domain assumption Lemma 2.1 and Lemma 2.2 from the authors' prior papers [13,14] are accepted as proved.
    These lemmas motivate the need for stencil extension and state the resolution effects of p and C_alpha. They are imported without proof in the present paper.
  • domain assumption Scale-independence requires tau and beta_k to have the same dimensions, [f]^2.
    Adopted in Section 2.2 as a design constraint for the scheme; it is a modeling choice, not a theorem.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Accuracy analysis and optimization of scale-independent third-order WENO-Z scheme with critical-point accuracy preservation." pith.science (2026). https://pith.science/paper/H6JWM3FZ

@misc{pith2026250908413,
  author       = {Pith},
  title        = {Pith review of: Accuracy analysis and optimization of scale-independent third-order WENO-Z scheme with critical-point accuracy preservation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/H6JWM3FZ}},
  note         = {Machine review of arXiv:2509.08413}
}
read the original abstract

To address the order degradation at critical points in the WENO3-Z scheme, some improvements have been proposed , but these approaches generally fail to consider the occurrence of critical points at arbitrary positions within grid intervals, resulting in their inability to maintain third-order accuracy when a first-order critical point (CP1) occurs. Also, most previous improved schemes suffer from a relatively large exponent p of the ratio of global to local smoothness indicators, which adversely affects the numerical resolution. Concerning these limitations, introduced here is an accuracy-optimization lemma demonstrating that the accuracy of nonlinear weights can be enhanced providing that smoothness indicators satisfy specific conditions, thereby establishing a methodology for elevating the accuracy of nonlinear weights. Leveraging this lemma, a local smoothness indicator is constructed with error terms achieving second-order in smooth regions and fourth-order at CP1, alongside a global smoothness indicator yielding fourth-order accuracy in smooth regions and fifth-order at CP1, enabling the derivation of new nonlinear weights that meet accuracy requirements even when employing p=1. Furthermore, a resolution-optimization lemma is proposed to analyze the relationship between parameters in local smoothness indicators and resolution. By integrating theoretical analysis with numerical practices, free parameters in non-normalized weights and local smoothness indicators are determined under the balance of numerical resolution and robustness, which leads to the development of WENO3-ZES4, a new WENO3-Z improvement that preserves the optimal order at CP1 especially with p=1. 1D and 2D validating tests show that the new scheme consistently achieves third-order in the case of CP1 regardless of its position and exhibits good resolution as well as preferable robustness.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages

  1. [1]

    Weighted essentially non-oscillatory schemes

    Liu X-D, Osher S, Chan T. Weighted essentially non-oscillatory schemes. J Comput Phys 1994;115:200–212

    work page 1994

  2. [2]

    Efficient implementation of weighted ENO schemes

    Jiang GS, Shu C-W. Efficient implementation of weighted ENO schemes. J Comput Phys 1996;126:202–228

    work page 1996

  3. [3]

    Mapped weigted essentially non-oscillatory schemes: achieving optimal order near critical points

    Henrick AK, Aslam TD, Powers JM. Mapped weigted essentially non-oscillatory schemes: achieving optimal order near critical points. J Comput Phys 2005;207:542–567

    work page 2005

  4. [4]

    An improved weighted essentially non -oscillatory scheme for hyperbolic conservation laws

    Borges R, Carmona M, Costa B, Don WS . An improved weighted essentially non -oscillatory scheme for hyperbolic conservation laws. J Comput Phys 2008;227:3191–3211

    work page 2008

  5. [5]

    High order weighted essentially no n-oscillatory WENO -Z schemes for hyperbolic conservation laws

    Castro M, Costa B , Don WS . High order weighted essentially no n-oscillatory WENO -Z schemes for hyperbolic conservation laws. J Comput Phys 2011;230:1766–1792

    work page 2011

  6. [6]

    Accuracy of the weighted essentially non -oscillatory conservative finite difference schemes

    Don WS, Borges R . Accuracy of the weighted essentially non -oscillatory conservative finite difference schemes. J Comput Phys 2013;250:347–372

    work page 2013

  7. [7]

    An improved WENO-Z scheme

    Acker F, Borges R, Costa B. An improved WENO-Z scheme. J Comput Phys 2016;313:726–753

    work page 2016

  8. [8]

    A high-resolution hybrid scheme for hyperbolic conservation laws

    Wu XS, Zhao YX. A high-resolution hybrid scheme for hyperbolic conservation laws. Int J Numer Meth Fluids 2015;78:162–187

    work page 2015

Show all 26 references
  1. [9]

    A new smoothness indicator for third-order WENO scheme

    Wu XS, Liang JH, Zhao YX. A new smoothness indicator for third-order WENO scheme. Int J Numer Meth Fluids 2016;81:451–459

    2016

  2. [10]

    Third-order WENO scheme with a new smoothness indicator

    Gande NR, Rathod Y , Rathan S. Third-order WENO scheme with a new smoothness indicator . Int J Numer Meth Fluids 2017;85:90–112

    2017

  3. [11]

    An improved third -order weighted essentially non-oscillatory scheme achieving optimal order near critical points

    Xu WZ, Wu WG. An improved third -order weighted essentially non-oscillatory scheme achieving optimal order near critical points. Comput Fluids 2018;162:113–125

    2018

  4. [12]

    Improvement of third -order WENO -Z scheme at critical points

    Xu WZ, Wu WG. Improvement of third -order WENO -Z scheme at critical points. Comput Math Appl 2018;75:3431–3452

    2018

  5. [13]

    Third-order scale-independent WENO-Z scheme achieving optimal order at critical points

    Li Q, Huang X, Yan P, Duan Y , You YC. Third-order scale-independent WENO-Z scheme achieving optimal order at critical points. Comput Fluids 2022;249:105703

    2022

  6. [14]

    On developing piecewise rational mapping with fine regulation capability for WENO schemes

    Li Q, Yan P, Huang X, Yang LQ, Xu F Y , You YC. On developing piecewise rational mapping with fine regulation capability for WENO schemes. J Sci Comput 2021;88:75

    2021

  7. [15]

    Improvements to Enhance Robustness of Third -Order 3 Scale-Independent WENO-Z Schemes

    Li Q, Yan P, Huang X. Improvements to Enhance Robustness of Third -Order 3 Scale-Independent WENO-Z Schemes. Adv Appl Math Mech 2024;17:373–406

    2024

  8. [16]

    Modified non -linear weights for fifth -order weighted essentially non-oscillatory schemes

    Kim CH, Ha Y , Yoon J. Modified non -linear weights for fifth -order weighted essentially non-oscillatory schemes. J Sci Comput 2016;67:299–323

    2016

  9. [17]

    An efficient class of WENO schemes with adaptive order

    Balsara DS, Garain S, Shu C -W. An efficient class of WENO schemes with adaptive order . J Comput Phys 2016;326:780–804

    2016

  10. [18]

    A family of high-order targeted ENO schemes for compressible-fluid simulations

    Fu L, Hu X-Y , Adams NA. A family of high-order targeted ENO schemes for compressible-fluid simulations. J Comput Phys 2016;305:333–359

    2016

  11. [19]

    Constructing higher order discontinuity-capturing schemes with upwind- biased interpolations and boundary variation diminishing algorithm

    Deng X, Shimizu Y , Xie B, Xiao F. Constructing higher order discontinuity-capturing schemes with upwind- biased interpolations and boundary variation diminishing algorithm. Comput Fluids 2020;200:104433

    2020

  12. [20]

    Li Q-J, Yokoi K, Xie Z-H, Omar S, Xue J -J. A fifth-order high-resolution shock-capturing scheme based on modified weighted essentially non -oscillatory method and boundary variation diminishing framework for compressible flows and compressible two-phase flows. Phys Fluids 202...

    2021

  13. [21]

    Affine-invariant WENO operator on nonuniform cartesian mesh with application to finite volume and discontinuous Galerkin methods

    Xiang XS, Wang BS, Gao Z. Affine-invariant WENO operator on nonuniform cartesian mesh with application to finite volume and discontinuous Galerkin methods. J Sci Comput 2025;102:2

    2025

  14. [22]

    Numerical simulation of Richtmyer-Meshkov instability of density-stratified interface of light-gas cylinder

    Cheng J-B, Fu D-X. Numerical simulation of Richtmyer-Meshkov instability of density-stratified interface of light-gas cylinder. Chin J Comput Phys 2001;18:390–396

    2001

  15. [23]

    The numerical simulation of two dimensional fluid flow with strong shocks

    Woodward P, Colella P. The numerical simulation of two dimensional fluid flow with strong shocks. J Comput Phys 1984;54:115–173

    1984

  16. [24]

    On positivity preserving high order discontinuous Galerkin schemes for compressible euler equations on rectangular meshes

    Zhang XX, Shu CW. On positivity preserving high order discontinuous Galerkin schemes for compressible euler equations on rectangular meshes. J Comput Phys 2010;229:8918–8934

    2010

  17. [25]

    Evaluation of tvd high resolution schemes for unsteady viscous shocked flows

    Daru V , Tenaud C. Evaluation of tvd high resolution schemes for unsteady viscous shocked flows . J Comput Fluid 2000;30:89–113

    2000

  18. [26]

    Direct simulation Monte Carlo simulations of hypersonic flows with shock interactions

    James NM, Bird GM. Direct simulation Monte Carlo simulations of hypersonic flows with shock interactions. AIAA J 2005;43:2565–2573

    2005

Pith tools

Reviewed August 4, 2026 · model on record in the stance chip above.